DEVELOPMENT OF A SHAPE MEMORY ALLOY WIRE ACTUATOR

Transcription

DEVELOPMENT OF A SHAPE MEMORY ALLOY WIRE ACTUATOR
JOURNAL OF THEORETICAL
AND APPLIED MECHANICS
52, 2, pp. 519-531, Warsaw 2014
DEVELOPMENT OF A SHAPE MEMORY ALLOY WIRE ACTUATOR TO
OPERATE A MORPHING WING
Misun Rim
KAIST, School of Mechanical, Aerospace and Systems Engineering, Daejeon, Korea
Eun-Ho Kim
University of South Carolina, Department of Mechanical Engineering, Columbia, USA
Woo-Ram Kang, In Lee
KAIST, School of Mechanical, Aerospace and Systems Engineering, Daejeon, Korea
e-mail: [email protected]
DSC tests were performed on several types of SMAs to verify the phase-transformation
temperatures, and then experiments to examine their characteristics were carried out. An
electric-current was supplied to the SMA wire to measure the appropriate operational current range. The force generated by the SMA wire increased according to the supplied current,
but it diminished when the over-current was supplied because thermo-mechanical properties
of the wire started to degrade. The appropriate stress range for effective actuation characteristics was also investigated. The SMA wire actuator was designed to operate a morphing
wing. Experiments for the wing were conducted to verify its characteristics and it was smoothly deformed.
Keywords: morphing wing, shape memory alloy (SMA), wire actuator
1.
Introduction
Aircraft characteristics should be optimized according to purposes and flight conditions (Joshi et
al., 2004; Bowman et al., 2007). Commercial passenger aircrafts are designed for a long distance
cruising, whereas combat aircrafts are designed for a short distance and a fast operation (Sanders
et al., 2003, 2005). Recent aerodynamic research has focused on active working systems which are
called smart structures for a variety of purposes or a need. A morphing wing is one type of smart
structures. Morphing can be classified in terms of two purposes. One is “configuration morphing”,
which entails changes in the span and the sweepback angle (Diaconu et al., 2008; Mattioni et al.,
2006). This morphing can lead to multi-mission capabilities and improved mobility difficult to
achieve with a conventional fixed wing. The other one is “maneuver morphing”, which creates a
controllable winglet to minimize vortices at the blade tip and vortices induced by flap changes.
This morphing can improve aerodynamic efficiency during operation by replacing the parts that
cause aerodynamic losses.
Most aircrafts use high-lift systems, which are connected by hinges, and they are passively
operated by more than one hydraulic actuator. For high aircraft efficiency, gaps between parts
causing drag and noise should be eliminated, and the structural complexity should be decreased.
Nowadays, there is increasing interest in replacing this mechanical operating method with a
comparatively light and simple morphing system (Cloyd, 1998; Thill et al., 2008).
Shape memory alloys (SMAs) have attractive characteristics such as the shape memory
effect and the superelasticity, because there are three phases in SMAs: martensite, austenite,
and R-phase. The martensite changes to the austenite when the temperature increases, and
the reverse occurs when the temperature decreases. Ms denotes the temperature at which the
520
M. Rim et al.
austenite starts to change to the martensite, and Mf denotes the temperature at which the
phase transformation is complete. As denotes the temperature at which the martensite starts
to change to the austenite, and Af denotes the temperature at which the phase transformation
is complete. The R-phase may or may not exist, depending on the manufacturing process for
the alloy. Under the phase-transformation temperature (Af ), the shape memory effect occurs
when the temperature is raised. The residual strains are fully recovered, and the original shape
is recovered at the phase-transformation temperature (Af ). Superelasticity occurs above the
phase-transformation temperature (Af ). The alloy returns to its initial elastic state after being
loaded and unloaded, and this process follows a hysteresis curve. These characteristics were
first discovered in 1951, and the shape memory effect was observed in 1963. Since then, SMAs
have become commercially available, and are used in many areas (Otsuka and Wayman, 1998).
Copper alloys and NiTi series (Nitinol) SMAs are commonly used alloys that exhibit the shape
memory effect (Duerig et al., 1999). Nitinol SMAs provide a great deal of energy per unit
volume, excellent corrosion resistance and biocompatibility, and long service lives, more than
100 000 cycles. Therefore, SMAs could be used for excellent actuators (Lee et al., 1999).
The SMA wire actuator was developed to control the shape of a morphing wing in this
research. A change in the camber of a morphing wing has the same effect as the operation of
a mechanical flap, while reducing the drag caused by the discontinuous parts of the flap so
that the aerodynamic characteristics can be significantly improved. Although the response of
the SMA wire actuator is slow, it was selected for this purpose because a fast response is not
required to operate a flap. Material property tests and actuation characteristic tests of the SMA
wire actuator were carried out, and Flexinol wire was selected as the most suitable actuator.
Developed actuators were used to operate a morphing flap, and actuation characteristic tests
of the morphing wing were performed. Matlab/Simulink was used to design the system which
is to investigate the response characteristics by measuring the displacement of the structures
deformed by the commanded current. The morphing wing was smoothly deformed when the
commanded current was supplied.
2.
2.1.
Experiments of the SMA wire actuator
DSC tests
Even the material properties of SMAs are already known, their precise values must be determined since these alloys exhibit complex nonlinear behavior, and their characteristics can be
altered easily by the manufacturing variables. Two types of Nitinol SMA wire were considered
as actuators for the morphing wing: SM495 (NDC) and Flexinol (Dynalloy, Inc.), which have
different compositions. Characteristic tests were conducted for each type to assess its suitability as an actuator. Table 1 lists the properties of each SMA and Table 2 lists diameters and
phase-transformation temperatures (Af ) of the wires, obtained from the product data sheet.
Differential scanning calorimeter (DSC) tests were carried out for two types of SMA wires
(SM495 and Flexinol) to determine phase-transformation temperatures (As , Af ). These tests
were only conducted for SM495 #1 and Flexinol #2 since different diameters of the same wire
had the same composition. Figure 1 shows results of these tests.
The SMA actuator is operated by changing its temperature, and the force is usually generated
when the martensite phase changes to the austenite phase. The measured phase-transformation
temperature Af was 67◦ C for SM495, and 80◦ C for Flexinol. These results differ slightly from
values of Table 2. The phase-transformation temperature of an SMA is related to its stress state,
and tends to increase with increasing applied stress. The R-phase appeared in all specimens
during DSC tests; this phase can vary according to the heat-treatment (annealing) temperature.
An SMA whose phase-transformation temperature exceeds room temperature will exhibit a
Development of a shape memory alloy wire actuator to operate a morphing wing
Table 1. Basic material properties of SMAs
–
–
–
–
–
–
–
–
–
–
–
SM495 (NDC)
Nickel: 54.5 wt%
Titanium: balance
Oxygen: max. 0.05 wt%
Carbon: max. 0.02 wt%
Melting point: 1310◦ C
Density: 6.5 g/cm3
Electrical resistivity: 76 µOhm cm
Modulus of elasticity: 28-41 · 103 MPa
Thermal expansion coefficient: 6.6 · 10−6 /◦ C
Ultimate tensile strength: 1100 MPa
Total elongation: 10%
–
–
–
–
–
–
–
Flexinol (Dynalloy, Inc.)
Thermal conductivity: 0.18 W/cm◦ C
Specific heat: 0.2 cal/g◦ C
Latent heat of transformation: 5.78 cal/g
Melting point: 1300◦ C
Density: 6.45 g/cm3
Electrical resistivity
Martensite: 80 µOhm cm
Austenite: 100 µOhm cm
Thermal expansion coefficient
Martensite: 6.6 · 10 − 6/◦ C
Austenite: 11.0 · 10 − 6/◦ C
Table 2. Types of SMA wire
SMA wire
Diameter [mm]
Af temperature [◦ C]
SM495 #1
SM495 #2
Flexinol #1
Flexinol #2
Flexinol #3
Flexinol #4
0.20
0.50
0.15
0.20
0.38
0.51
∼ 75
∼ 75
70
70
70
70
Fig. 1. Results of DSC tests: (a) SM495 #1, (b) Flexinol #2
shape memory effect at room temperature, and can be used as an actuator. Therefore, SM495
and Flexinol are both appropriate for use as actuators.
2.2.
Actuation characteristic tests
The temperature of a SMA wire varies when it is used as an actuator. Typically, the temperature is increased by electric heating and decreased by natural cooling, such as convection and
radiation. When a SMA wire is heated electrically, it is important to determine the appropriate
voltage or current to maintain the temperature within the operational range. Electric heating
raises the temperature very quickly, and this sometimes leads to over-current. Over-current causes the heat treatment (annealing) of a SMA wire actuator and the material properties change.
As a result, the characteristics of the actuator also change. Accordingly, several experiments
were carried out to measure the appropriate operational current for the SMA wire actuator.
521
522
M. Rim et al.
2.2.1.
Testing procedure
The length of each SMA wire was 10 cm, and its ends were fixed (Fig. 2). The force was
measured by a load cell at one end of the wire while a constant current was supplied to the wire.
A square-wave current was applied three times, and was increased in 0.1 A increments from 0.3 A
to 1.2 A in order to determine the appropriate current.
Fig. 2. Experimental setup
2.3.
SM495 wire
The force generated by SM495 wire varies according to the initial strain (pre-strain). Therefore, experiments were performed under different initial conditions (1%, 2%, 3%, and 4% of
pre-strain) to investigate the characteristics in terms of pre-strain. Figure 3 shows results obtained by measuring the force under each pre-strain condition as the current was increased
gradually. The generated force increased with increasing pre-strain, and also with the increasing
current up to about 0.6 A in all cases. The maximum force was generated at 0.6-0.7 A. Above
0.8 A, large force variations began to appear, and the force decreased as the current increased.
At low currents, the temperature increase was inadequate, and the force generated by shrinkage
caused by the shape memory effect of the SMA wire was low. On the other hand, at over-currents,
the temperature increase was excessive, the properties of the wire were changed, and the force
was diminished again. Therefore, an applied current under 0.5 A did not raise the temperature
of the wire sufficiently to generate the maximum force. At currents over 0.8 A, the temperature
increase was excessive and properties of the wire were altered, resulting in material degradation. Based on these results, 0.6 A was considered the appropriate current for the SM495 #1
(diameter: 0.2 mm) specimen.
Table 3. Response characteristics of the SM495 #1 wire actuator
SM495 wire
1%
2%
3%
4%
pre-strain
pre-strain
pre-strain
pre-strain
Frequency
[Hz]
0.2
0.17
0.13
0.12
Max. force
[N]
15
19
21
22
Power for
10 cm length
1.26 W
1.26 W
1.26 W
1.26 W
Development of a shape memory alloy wire actuator to operate a morphing wing
Fig. 3. Generated force by SM495 #1 (diameter: 0.2 mm) according to the current; (a) 1% pre-strain,
(b) 2% pre-strain, (c) 3% pre-strain, (d) 4% pre-strain
Fig. 4. (a) Generated force by SM495 #1 at 0.6 A. (b) Response time of SM495 #1
As shown in Fig. 3, the generated force gradually decreased when a 0.6 A current was applied
three times. The reason for this was due to phase reorientation that occurred to stabilize the
internal stress when the current started to flow. The 0.6 A current was applied repeatedly until
the force converged. Figure 4a shows the force under repeated applications of the current in
the 1% pre-strain case. The force generated by the wire gradually decreased as the number of
applications of the current increased, and converged after 20 cycles in all cases. The response
time according to the square wave current was measured when the force converged. Specifically,
the response time was investigated by the sum of two periods; the time until the force reached
90% of its maximum when the current was supplied, and the time until the force decreased to
10% of its maximum when the current was cut off (Fig. 4b). Table 3 lists the response frequencies,
maximum forces, and required power for the 10-cm SM495 #1 wire specimen according to the
pre-strain.
2.3.1.
Flexinol wire
Variation of the generated force according to the pre-strain was not observed for the Flexinol
wire (Fig. 5). Therefore, the response characteristics of Flexinol were investigated in the absence
of the pre-strain; the results are presented in Table 4 and Fig. 6.
523
524
M. Rim et al.
Fig. 5. Generated force by Flexinol #1 (diameter: 0.15 mm) according to the pre-strain at 0.5 A;
(a) 1% pre-strain, (b) 2% pre-strain, (c) 3% pre-strain, (d) 4% pre-strain
Table 4. Response characteristics of Flexinol
Flexinol wire
Flexinol
Flexinol
Flexinol
Flexinol
#1
#2
#3
#4
Current
[A]
0.5
0.7
1.7
2.4
Max. force
[N]
15
25
95
145
Frequency
[Hz]
0.20
0.18
0.07
0.04
Power for
10 cm length
1.55 W
1.47 W
3.91 W
3.80 W
Fig. 6. Results of actuation characteristic tests of Flexinol; (a) appropriate current, (b) maximum force,
(c) response frequency
The appropriate current and maximum force for Flexinol increased quadratically with respect
to the diameter. The reason for this was the constant stress-strain relationship for the given
material, so that the cross-sectional area of the wire was proportional to the required current
and the force. The response frequency was inversely proportional to the wire diameter. The
volume per unit length increased quadratically with respect to the wire diameter, and thus
heating and cooling required more time because of the increased heat capacity.
As a linear actuator, the Flexinol wire is more advantageous than the SM495 wire because
the generated force does not vary appreciably according to the pre-strain.
Development of a shape memory alloy wire actuator to operate a morphing wing
2.4.
Stress-strain tests of Flexinol wires
Since the deformation of an SMA wire varies according to the applied force, it is necessary to
determine the appropriate operational range of forces for each type of the wire in order to design
an actuator. Therefore, stress measurement experiments were performed for the four types of
Flexinol wires. A constant force was applied to the Flexinol wire, and the displacement of the
end of the wire was measured to determine the deformation under periodic application of the
current listed in Table 4. Weights were suspended from the end of the wire to apply the constant
force (Fig. 7). A laser displacement sensor was used to measure the displacement of the end of
the wire. The displacements obtained when the constant force was applied to the Flexinol #1
wire are shown in Table 5. The initial length of the wire was 300 mm. The wire contracted
when it was heated by the applied current, and was stretched by the suspended weights when it
cooled while the current was cut off. However, the amount of extension during cooling gradually
decreased when the force applied by the suspended weights was insufficient (0.27 N, 0.54 N, and
1.03 N). These results show that if the force applied to the wire is inadequate, the wire cannot
be recovered to its initial shape after deformation. When the applied force was appropriate
(2.02 N, 2.98 N, and 3.94 N), the amount of the deformation was constant, and the wire was fully
recovered to its initial length when the current was repeatedly supplied and cut off. On the other
hand, when the excessive force was applied (6.87 N, 7.85 N, and 9.81 N), the wire was stretched
beyond its initial length as the current was repeatedly applied. In the 6.87 N and 7.85 N cases,
the amount of extension during cooling converged as the number of current cycles increased.
This means that the characteristics of the SMA wire actuator change when the excessive force
is applied, but the behavior of the actuator is stabilized after repetitive operation. When 9.81 N
was applied, the length of the wire increased with the number of current cycles, and the wire
was finally broken.
Fig. 7. Experimental setup for measuring the strain on a Flexinol wire
Figure 8a shows the strain of the Flexinol #1 wire according to the applied force for the
applied currents of 0.3, 0.4, and 0.5 A. The strain of the wire decreased at the currents lower
than the appropriate current, and was about 5% at the appropriate current of 0.5 A when a 3-7 N
force was applied. From these results, the appropriate range of the applied force per wire was
3-5 N when the actuator was designed using the Flexinol #1 wire. Figure 8b shows the strain
according to the applied stress for the four types of Flexinol wires; all of them exhibited similar
tendencies. The measured appropriate stress range for effective operation was 200-500 MPa.
525
526
M. Rim et al.
Table 5. Deformation of the SMA Flexinol #1 wire according to the applied force
Force
0.27 N
0.54 N
1.03 N
2.02 N
2.98 N
3.94 N
6.87 N
7.85 N
9.81 N
Displacement of
the end of
the wire
Force
Displacement of
the end of
the wire
Force
Displacement of
the end of
the wire
Fig. 8. Strain on the Flexinol #1 wire (a) and strain for the four types of Flexinol wires (b)
2.5.
Response of Flexinol wire actuators to a commanded input current
The system used to investigate the response characteristics of the SMA wire actuator was designed via Matlab/Simulink. This system measured the displacement response to a commanded
input current. The Flexinol wire with a diameter of 0.20 mm (Flexinol #2) was selected as the
most suitable actuator among the four types of Flexinol wires since its response was relatively
fast and it generated a large force. A constant force was applied by suspending a mass of 650 g
from the end of the actuator. The commanded input current was supplied to the actuator by the
designed control algorithm system. An additional current amplifier was included in this system
because the output current of the data acquisition (DAQ) board was insufficient to operate the
actuator. The amplified current operated the actuator, and the length of the actuator changed
accordingly. The laser displacement sensor measured the displacement of the end of the actuator
to determine the change in the length of the actuator. The DAQ board collected data, and they
were stored in the personal computer (PC) memory (Fig. 9).
Development of a shape memory alloy wire actuator to operate a morphing wing
Fig. 9. System to measure the displacement response to a commanded input current
Fig. 10. Deformation of the SMA wire actuator according to the square wave input current; length
change at: (a) 300 mA, (b) 400 mA, (c) 500 mA, (d) 600 mA
The experiments were performed for four different commanded currents: square wave currents
of 300, 400, 500, and 600 mA. In these tests, it took more time when the commanded current was
cut off than when the commanded current was applied. The output displacement was relatively
constant in response to the 300, 400, and 500 mA command input currents (Fig. 10). As noted
in Section 2.2.3, if the current supplied to the SMA wire actuator is more than 700-800 mA, the
material properties are degraded, and the length of the SMA wire is gradually extended as the
number of current cycles increases. When the input current was 600 mA, the length of the SMA
wire actuator was extended (Fig. 10d). The one possible reason for this phenomenon could be
concerned as that the additional current amplifier operated as an independent system when the
commanded current was applied and it caused an overshoot. This overshoot may have caused the
current to exceed 700 mA so that the properties of the SMA were degraded. Therefore, a control
527
528
M. Rim et al.
algorithm to cut the current when it exceeds the critical current level should be developed and
added to the system.
The output displacement was measured under an increasing step commanded current
(0 mA →100 mA →200 mA →300 mA→400 mA→500 mA→400 mA→300 mA→200 mA→100 mA
→0 mA) (Fig. 11). It was not proportional to the input current due to the nonlinearity of the
material properties of the SMA. This output displacement jumped sharply at the 300 mA step
when the current was increasing, and dropped sharply at the 100 mA step when the current
was decreasing. The cause of this phenomenon was attributed to SMA phase transformation,
which indicates that hysteretic behavior occurred when the current was applied.
Fig. 11. Deformation of the SMA wire actuator according to the increasing/decreasing input current
3.
3.1.
Experiments on the morphing wing
Actuation characteristic tests of the morphing wing
Conventional mechanical hinges, such as those that connect the flaps to the main wings, cause aerodynamic loss. This problem can be improved by using a morphing wing that provides the
same functionality without hinges. The morphing wing model developed by Kang et al. (2012)
was used in this research. The current was supplied to the SMA wire actuator installed in the
morphing wing model to operate it, and the flap rotated smoothly. The applied current was
increased to investigate the relationship between the current and the flap angle. The angular
displacement of the flap increased linearly in the range 1.5-2.7 A, but exhibited nonlinear behavior when the current exceeded 2.7 A. The reason for this appears to be that excessive applied
current caused heat-treatment, and the properties of the SMA wire were degraded.
According to the results of an aerodynamic analysis of the morphing wing at a flight velocity
20 m/s, the difference between the pressure coefficients of the upper and lower skins increased
as the flap angle increased from 0◦ to 6.4◦ to 15.9◦ . The highest lift-to-drag ratio was estimated
at 15.22 when the flap angle was 6.4◦ ; this was because both the lift and drag coefficients
increased when the flap angle increased (Kang et al., 2012). This means, it is necessary to
determine the most efficient flap angle to improve the aerodynamic efficiency, instead of simply
increasing the flap angle unconditionally.
3.2.
Response to a commanded input current
The system used to investigate the response characteristics of the morphing wing was designed via Matlab/Simulink. This system was used to measure the displacement response to a
commanded input current. The Flexinol wire with a diameter of 0.20 mm (Flexinol #2) was used
as the actuator of the morphing wing. The input current (controlled by the PC) was supplied by
the DAQ board and amplified by the current amplifier to operate the morphing wing (Fig. 12).
To verify the deformation, the displacement of the trailing edge of the flap was measured instead
of the flap rotation angle.
Development of a shape memory alloy wire actuator to operate a morphing wing
Fig. 12. System to measure the displacement response to a commanded input current
The time required for the displacement to reach 80% of the converged value was recorded as the response time. The response time was 10.34 s when a constant current of 2 A
was applied, and 3.8 s when the current was cut off. The response was faster when the current was cut off than when the current was applied, which is contrary to the results obtained for the SMA wire. The reason can be considered as that the elastic force of the morphing wing and the additional rubber ring shortened the response time when the current
was cut off. When the increasing/decreasing step input current (0 mA→500 mA→1000 mA
→1500 mA→2000 mA→1500 mA→1000 mA→500 mA→0 mA) was applied, large variations of
the displacement occurred at 1000, 1500, and 2000 mA. The nonlinear behavior of the SMA wire
during the phase transformation could be the reason for this phenomenon (Fig. 13).
Fig. 13. Displacement of the trailing edge of the morphing wing; (a) constant current of 2 A, (b) square
wave current of 2 A, (c) increasing/decreasing current
529
530
M. Rim et al.
4.
Conclusion
The characteristics of the SMA wire actuator and the morphing wing operated by this actuator
were investigated in this research. DSC tests were carried out to confirm the exact material
properties of SMAs, such as the phase-transformation temperature. SM495 and Flexinol both
exhibited the shape memory effect at room temperature, and hence both were deemed suitable
for use as actuators. Several actuation characteristic tests were performed on the SMA wire
actuators, and the appropriate currents were measured. For the SM495 wire with a diameter of
0.2 mm (SM495 #1), the generated force increased with the pre-strain. A current less than 0.5 A
was insufficient to fully heat this wire, and a current over 0.8 A was excessive, causing degradation
of the SMA properties. Consequently, 0.6-0.7 A was considered appropriate for operating this
SMA wire actuator. For Flexinol, on the other hand, significant variation of the generated
force according to the pre-strain was not observed, and thus the response characteristics were
investigated without the pre-strain. The appropriate current and the maximum generated force
increased quadratically with respect to the diameter. Flexinol was selected as a more suitable
actuator because it was less affected by the pre-strain than SM495. Since the deformation of an
SMA varies according to the applied force, it was necessary to investigate the appropriate range
of applied forces. Therefore, stress-strain tests were conducted for the four types of Flexinol wires
with different diameters considered in this study. When a current lower than the appropriate
current was supplied to the wire, the strain decreased. The maximum strain of 5% appeared
when 200-500 MPa of stress was applied to the wire at the appropriate current. This stress range
was considered to be the appropriate range; all four types of Flexinol wires exhibited similar
tendencies. Based on the results of these experiments, Flexinol #2, which had a diameter of
0.2 mm, was selected as the actuator to operate the morphing wing. The system was designed
via Matlab/Simulink to investigate the response characteristics of the SMA wire actuator and
morphing wing structures to the commanded input current. The SMA wire actuator and the
morphing wing deformed smoothly, but also exhibited nonlinear behavior due to the phase
transformation of the SMA at certain current levels. This phenomenon should be considered
when the SMA wire actuator is designed for use in structures such as morphing wings.
Acknowledgements
This research was supported by the KARI-University Partnership Program. This research was also
partly supported by WCU (World Class University) program through the National Research Foundation
of Korea funded by the Ministry of Education, Science and Technology (R31-2008-000-10045-0).
References
1. Bowman J., Sanders B., Cannon B., Kudva J., Joshi S., Weisshaar T., 2007, Development of next generation morphing aircraft structures, Proc. 48th AIAA/ASME/ASCE/AHS/ASC
Structures, Structural Dynamics and Materials Conf. (SDM), AIAA-2007-1730
2. Cloyd J.S., 1998, Status of the United States Air Force’s more electric aircraft initiative, IEEE
Aerospace and Electronic Systems Magazine, 13, 4, 17-22
3. Diaconu C.G., Weaver P.M., Mattioni F., 2008, Concepts for morphing airfoil sections using
bi-stable laminated composite structures, Thin-Walled Structures, 46, 6, 689-701
4. Duerig T., Pelton A., Stockel D., 1999, An overview of nitinol medical applications, Materials
Science and Engineering, A, 273/275, 149-160
5. Joshi S.P., Tidwell Z., Crossley W.A., Ramakrishnan S., 2004, Comparison of morphing wing strategies based upon aircraft performance impacts, Proc. 45th
AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials Conf. (SDM),
AIAA-2004-1722
Development of a shape memory alloy wire actuator to operate a morphing wing
6. Kang W.R., Kim E.H., Jeong M.S., Lee I., 2012, Morphing wing mechanism using a SMA
wire actuator, International Journal of Agricultural and Soil Science, 13, 1, 58-63
7. Lee H.J., Lee J.J., Huh J.S., 1999, A simulation study on the thermal buckling behavior of laminated composite shells with embedded shape memory alloy (SMA) Wires, Composite Structures,
47, 1/4, 463-469
8. Mattioni F., Gatto A., Weaver P.M., Friswell M.I., Potter K.D., 2006, The application of residual stress tailoring of snap-through composites for variable sweep wings, Proc. 47th
AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conf. (SDM),
AIAA-2006-1972
9. Otsuka K., Wayman C.M., 1998, Shape Memory Materials, Cambridge University Press
10. Sanders B., Estep F.E., Forster E., 2003, Aerodynamic and aeroelastic characteristics of
wings with conformal control surfaces for morphing aircraft, Journal of Aircraft, 40, 1, 94-99
11. Sanders B., Reich G.W., Joo J.J., 2005, Conceptual skin design for morphing aircraft, Proc.
16th Int. Conf. on Adaptive Structures and Technologies (ICAST), 16, 275-281
12. Thill C., Etches J., Bond I., Potter K., Weaver P., 2008, Morphing skins, Aeronautical
Journal, 112, 1129, 117-139
Manuscript received August 8, 2013; accepted for print December 1, 2013
531